《人工智能 英文版》求取 ⇩

1 Introduction1

1．1 What Is AI?1

1．2 Approaches to Artificial Intelligence6

1．3 Brief History of AI8

1．4 Plan of the Book11

1．5 Additional Readings and Discussion14

Exercises17

ⅠReactive Machines19

2 Stimulus-Response Agents21

2．1 Perception and Action21

2．1．1 Perception24

2．1．2 Action24

2．1．3 Boolean Algebra25

2．1．4 Classes and Forms of Boolean Functions26

2．2 Representing and Implementing Action Functions27

2．2．1 Production Systems27

2．2．2 Networks29

2．2．3 The Subsumption Architecture32

2．3 Additional Readings and Discussion33

Exercises34

3Neural Networks37

3．1 Introduction37

3．2 Training Single TLUs38

3．2．1 TLU Geometry38

3．2．2 Augmented Vectors39

3．2．3 Gradient Descent Methods39

3．2．4 The Windrow-Hoff Procedure41

3．2．5 The Generalized Delta procedure41

3．2．6 The Error-Correction Procedure43

3．3 Neural Networks44

3．3．1 Motivation44

3．3．2 Notation45

3．3．3 The Backpropagation Method46

3．3．4 Computing Weight Changes in the Final Layer48

3．3．5 Computing Changes to the Weights in Intermediate Layers48

3．4 Generalization ,Accuracy, and Overfitting51

3．5 Additional Readings and Discussion54

Exercises55

4 Machine Evolution59

4．1 Evolutionary Computation59

4．2 Genetic Programming60

4．2．1 Program Representation in GP60

4．2．2 The GP Process62

4．2．3 Evolving a Wall-Following Robot65

4．3 Additional Readings and Discussion69

Exercises69

5State Machines71

5．1 Representing the Environment by Feature Vectors71

5．2 Elman Networks73

5．3 Iconic Representations74

5．4 Blackboard Systems77

5．5 Additional Readings and Discussion80

Exercises80

6 Robot Vision85

6．1 Introduction85

6．2 Steering and Automobile86

6．3 Two Stages of Robot Vision88

6．4 Image Processing91

6．4．1 Averaging91

6．4．2 Edge Enhancement93

6．4．3 Combining Edge Enhancement with Averaging96

6．4．4 Region Finding97

6．4．5 Using Image Attributes Other Than Intensity101

6．5 Scene Analysis102

6．5．1 Interpreting Lines and Curves in the Image103

6．5．2 Model -Based Vision106

6．6 Stereo Vision and Depth Information108

6．7 Additional Readings and Discussion110

Exercises111

Ⅱ Search in State Spaces115

7 Agents That Plan117

7．1 Memory Versus Computation117

7．2 State-Space Graphs118

7．3 Searching Explicit State Spaces121

7．4 Feature-Based State Spaces122

7．5 Graph Notation124

7．6 Additional Readings and Discussion125

Exercises126

8 Uninformed Search129

8．1 Formulating the State Space129

8．2 Components of Implicit State-Space Graphs130

8．3 Breadth-First Search131

8．4 Depth-First or Backtracking Search133

8．5 Iterative Deepening135

8．6 Additional Readings and Discussion136

Exercises137

9 Heuristic Search139

9．1 Using Evaluation Functions139

9．2 A General Graph-Searching Algorithm141

9．2．1 Algorithm A142

9．2．2 Admissibility of A145

9．2．3 The Consistency (or Monotone)Condition150

9．2．4 Iterative-Deepening A153

9．2．5 Recursive Best-First Search154

9．3 Heuristic Functions and Search Efficiency155

9．4 Additional Readings and Discussion160

Exercises160

10 Planning, Acting ,and Learning163

10．1 The Sense/Plan/Act Cycle163

10．2 Approximate Search165

10．2．1 Island-Driven Search166

10．2．2 Hierarchical Search167

10．2．3 Limited-Horizon Search169

10．2．4 Cycles170

10．2．5 Building Reactive Procedures170

10．3 Learning Heuristic Functions172

10．3．1 Explicit Graphs172

10．3．2 Implicit Graphs173

10．4 Rewards Instead of Goals175

10．5 Additional Readings and Discussion177

Exercises178

11Alternative Search Formulations and Applications181

11．1 Assignment Problems181

11．2 Constructive Methods183

11．3 Heuristic Repair187

11．4 Function Optimization189

Exercises192

12 Adversarial Search195

12．1 Two-Agent Games195

12．2 The Minimax Procedure197

12．3 The Alpha-Beta Procedure202

12．4 The Search Efficiency of the Alpha-Beta Procedure207

12．5 Other Important Matters208

12．6 Games of Chance208

12．7 Learning Evaluation Functions210

12．8 Additional Readings and Discussion212

Exercises213

Ⅲ Knowledge Representation and Reasoning215

13The Propositional Calculus217

13．1 Using Constraints on Feature Values217

13．2 The Language219

13．3 Rules of Inference220

13．4 Definition of Proof221

13．5 Semantics222

13．5．1 Interpretations222

13．5．2 The Propositional Truth Table223

13．5．3 Satisfiability and Models224

13．5．4 Validity224

13．5．5 Equivalence225

13．5．6 Entailment225

13．6 Soundness and Completeness226

13．7 The PSAT Problem227

13．8 Other Important Topics228

13．8．1 Language Distinctions228

13．8．2 Metatheorems228

13．8．3 Associative Laws229

13．8．4 Distributive Laws229

Exercises229

14 Resolution in the Propositional Calculus231

14．1 A New Rule of Inference: Resolution231

14．1．1 Clauses as wffs231

14．1．2 Resolution on Clauses231

14．1．3 Soundness of Resolution232

14．2 Converting Arbitrary wffs to Conjunctions of Clauses232

14．3 Resolution Refutations233

14．4 Resolution Refutation Search Strategies235

14．4．1 Ordering Strategies235

14．4．2 Refinement Strategies236

14．5 Horn Clauses237

Exercises238

15 The Predicate Calculus239

15．1 Motivation239

15．2 The Language and Its Syntax240

15．3 Semantics241

15．3．1 Worlds241

15．3．2 Interpretations242

15．3．3 Models and Related Notions243

15．3．4 Knowledge244

15．4 Quantification245

15．5 Semantics of Quantifiers246

15．5．1 Universal Quantifiers246

15．5．2 Existential Quantifiers247

15．5．3 Useful Equivalences247

15．5．4 Rules of Inference247

15．6 Predicate Calculus as a Language for Representing Knowledge248

15．6．1 Conceptualizations248

15．6．2 Examples248

15．7 Additional Readings and Discussion250

Exercises250

16Resolution in the Predicate Calculus253

16．1 Unification253

16．2 Predicate-Calculus Resolution256

16．3 Completeness and Soundness257

16．4 Converting Arbitrary wffs to Clause Form257

16．5 Using Resolution to Prove Theorems260

16．6 Answer Extraction261

16．7 The Equality Predicate262

16．8 Additional Readings and Discussion265

Exercises265

17Knowledge-Based Systems269

17．1 Confronting the Real World269

17．2 Reasoning Using Horn Clauses270

17．3 Maintenance in Dynamic Knowledge Bases275

17．4 Rule-Based Expert Systems280

17．5 Rule Learning286

17．5．1 Learning Propositional Calculus Rules286

17．5．2 Learning First-Order Logic Rules291

17．5．3 Explanation-Based Generalization295

17．6 Additional Readings and Discussion297

Exercises298

18 Representing Commonsense Knowledge301

18．1 The Commonsense World301

18．1．1 What Is Commonsense Knowledge?301

18．1．2 Difficulties in Representing Commonsense Knowledge303

18．1．3 The Importance of Commonsense Knowledge304

18．1．4 Research Areas305

18．2 Time306

18．3 Knowledge Representation by Networks308

18．3．1 Taxonomic Knowledge308

18．3．2 Semantic Networks309

18．3．3 Nonmonotonic Reasoning in Semantic Networks309

18．3．4 Frames312

18．4 Additional Readings and Discussion313

Exercises314

19Reasoning with Uncertain Information317

19．1 Review of Probability Theory317

19．1．1 Fundamental Jdeas317

19．1．2 Conditional Probabilities320

19．2 Probabilistic Inference323

19．2．1 A General Method323

19．2．2 Conditional Independence324

19．3 Bayes Networks325

19．4 Patterns of Inference in Bayes Networks328

19．5 Uncertain Evidence329

19．6 D-Separation330

19．7 Probabilistic Inference in Polytrees332

19．7．1 Evidence Above332

19．7．2 Evidence Below334

19．7．3 Evidence Above and Below336

19．7．4 A Namerical Example336

19．8 Additional Readings and Discussion338

Exercises339

20Learning and Acting with Bayes Nets343

20．1 Learning Bayes Nets343

20．1．1 Known Network Structure343

20．1．2 Learning Networks Structure346

20．2 Probabilistic Inference and Action351

20．2．1 The General Setting351

20．2．2 An Extended Example352

20．2．3 Generalizing the Example356

20．3 Additional Readings and Discussion358

Exercises358

Ⅳ Planning Methods Based on Logic361

21The Situation Caluclus363

21．1 Reasoning about States and Actions363

21．2 Some Difficulties367

21．2．1 Frame Axioms367

21．2．2 Qualifications369

21．2．3 Ramifications369

21．3 Generating Plans369

21．4 Additional Readings and Discussion370

Exercises371

22 Planning373

22．1 STRIPS Planning Systems373

22．1．1 Describing States and Goals373

22．1．2 Forward Search Methods374

22．1．3 Recursive STRIPS376

22．1．4 Plans with Run-Time Conditionals379

22．1．5 The Sussman Anomaly380

22．1．6 Backward Search Methods381

22．2 Plan Spaces and Partial-Order Planning385

22．3 Hierarchical Planning393

22．3．1 ABSTRIPS393

22．3．2 Combining Hierarchical and Partial-Order Planning395

22．4 Learning Plans396

22．5 Additional Readings and Discussion398

Exercises400

ⅤCommunication and Integration405

23Multiple Agents407

23．1 Interacting Agents407

23．2 Models of Other Agents408

23．2．1 Varieties of Models408

23．2．2 Simulation Strategies410

23．2．3 Simulated Databases410

23．2．4 The Intentional Stance411

23．3 A Modal Logic of Knowledge412

23．3．1 Modal Operators412

23．3．2 Knowledge Axioms413

23．3．3 Reasoning about Other AgentsKnowledge415

23．3．4 Predicting Actions of Other Agents417

23．4 Additional Readings and Discussion417

Exercises418

24Communication among Agents421

24．1 Speech Acts421

24．1．1 Planning Speech Acts423

24．1．2 Implementing Speech Acts423

24．2 Understanding Language Strings425

24．2．1 Phrase-Structure Grammars425

24．2．2 Semantic Analysis428

24．2．3 Expanding the Grammar432

24．3 Efficient Communication435

24．3．1 Use of Context435

24．3．2 Use of Knowledge to Resolve Ambiguities436

24．4 Natural Language Processing437

24．5 Additional Readings and Discussion440

Exercises440

25 Agent Architectures443

25．1 Three-Level Architectures444

25．2 Goal Arbitration446

25．3 The Triple-Tower Architecture448

25．4 Bootstrapping449

25．5 Additional Readings and Discussion450

Exercises450

Bibliography453

Index493